AI Research Radar
July 23–29: 12 frontier papers out of 1,847 read.
Frontier
Papers rated as moving the field
How we chooseKimi K3
“Together with Stable LatentMoE, which effectively activates 16 of 896 routed experts per token, and refined training and data recipes, these advances yield an approximately 2.5x improvement in overall scaling efficiency over Kimi K2.”
Profile-Graph Memory for LLM Agents
“ProGraph averages 80.1% on MemHop (matching the FullContext reference) and 78.4% on LoCoMo (exceeding FullContext by 11.3pp), outperforming Mem0, A-Mem, HippoRAG, and RAG on both.”
DARWIN
“DARWIN-Guard achieves an average unsafe recall of 91.6% across 12 safety benchmarks, outperforming strong guardrails such as YuFeng-XGuard and Nemotron Guard, while maintaining a nearly 100% pass rate on standard benign datasets.”
TLA⁺-Bench
ITPEval
NavVerse
“Zero-shot experiments with RL, VLA, and modular baselines show that current agents remain far from solving cross-context navigation: end-to-end VLAs obtain the highest zero-shot success, while the modular method provides the strongest safety profile.”
MV-Bench
Explanation-Bound Tool Execution for AI Agents
Architectural Backdoors in Vision-Language Model Supply Chains via Representation Steering
“The results show that the proposed architectural steering backdoor compromises integrity, safety enforcement, and ranking fairness while preserving normal behavior on clean inputs.”
SAGE
Solar Open 2 Technical Report
The week by field
Jul 23–29: 171 cleared, 12 frontier
The week’s papers have not been sorted into fields yet.
Models under test
Papers this week that name the model
Labs publishing
Papers this week with an author at the lab
Also cleared
159 papers
- Harm is not Universal: Community-Specific Toxicity Detection is Urgently NeededMicrosoft
- Mage-VL: An Efficient Codec-Native Streaming Multimodal Foundation ModelMicrosoft
- The User Asks, Platforms Compete: How Agentic Recommendation Markets Take ShapeTencent
- Hybrid Analysis for Secure MCP Tool Use in LLM AgentsAlibaba
- Qwen-Audio-3.0-TTS: Freely Controllable and Highly Robust Speech Synthesis with Multi-Stage Training ParadigmAlibaba
- FilmBench: A Film-Grade Benchmark for Cinematic Video GenerationAlibaba
- Do VLMs Read or Rewrite? On Transcription Faithfulness in Vision-Language ModelsAmazon
- JAXBench: Benchmarking Autonomous TPU Kernel OptimizationGoogle
- Windowed-MTP: Removing the Full-Context Draft-KV Tax at Million-Token ContextNVIDIA
- OpenForgeRL: Train Harness-native Agents in Any EnvironmentMicrosoft
- FORCE-Bench: A Benchmark, Dataset, and Evaluation Harness for Agentic AI in Enterprise FinanceMicrosoft
- Efficient Clustering with Provable Guardrails for LLM Inference at ScaleAmazon
All 159 papers
- Ranked by Position: Order Sensitivity as an Exploitable Attack Surface in LLM Listwise RecommendersCode
- ScalableRAG: High-Quality RAG at Zero Ingestion CostCode
- LabRobFail: A Benchmark for Robotic Failure Analysis in Chemical Self-driving LaboratoriesCode
- PhantomSeal: Proactive Deepfakes Defense with Identity/Context Protection and Forensic TracingCode
- Explainable Deepfake Detection ChallengeCode
- When Are Reasoning-Based Guardrails Not Efficient? ResponseGuard: A Fast Vision-Language Guard for Real-Time ModerationCode
- FineServe: A Fine-Grained Dataset and Characterization of Global LLM Serving WorkloadsCode
- JailMeter: An Evidence-Based Evaluation Framework for Jailbreak Attacks on Large Language ModelsCode
- BaseRT: Advancing Best-in-Class LLM Inference with Apple M5 Neural AcceleratorsCode
- GhostPrompt: Cross-Image Adversarial Prompt for Vision-Language ModelsCode
- Silent Failures in Multimodal Agentic Search:A Diagnostic Taxonomy and Cross-Judge EvaluationCode
- RDQ: Residual Distribution Quantization for Large Language Models
- ContainmentBench: Trace-Based Evaluation of Post-Injection Containment in Tool-Using LLM Agents
- LLM Scheming Inversely Scales with Pretraining Language Coverage
- The Mirage of LLM Guardrails: A Case Study in AI-Assisted Medical Note Manipulation
- The Missing Layer: Specification Infrastructure for AI Oversight
- Early Detection of Distributed Backdoors in Multi-Agent LLM Systems: A Characterization Study
- TYPO: Instruction-Dense Visual Jailbreaks against Commercial Closed-Source Image-Generation Models
- ALIBI: Adaptive Agentic Attacks on LLM-Based Vulnerability Detectors via Adversarial Code Comments
- Decision-Level Hijacking: Injecting Cognitive Bias into Large Language Models via Bit-Flip Attacks
- Instruction-Tuned Language Models Cannot Sample from Distributions They Can Describe
- Cyber-Capable AI Agents: Vulnerabilities, Evaluation Containment, and Defensive Response
- From Dyad to Triad: Eliciting XAI Requirements in Stroke Rehabilitation
- I2VShield: An Efficient Proactive Defense Framework against DiT-based Image-to-Video Models
- PILA: Plug-and-Play Insertion for LLM-native Advertising
- SkillGate: Cost Efficient Runtime Malicious Skill File Detection in Coding Agents
- Construction-Driven Injection: Linguistically-Grounded Edit-Based Code-Mixing Fingerprints for Large Language Models
- Why Public Service AI Governance Frameworks Risk Failing in the Age of General-Purpose AI: Lessons from Policing
- Runtime Uncertainty Monitoring for LLM-Based Multi-Agent Systems Using Bayesian Networks
- TIGA: Trajectory-Injected Generative Attack against Black-box AIGC Detectors
- Minimizing Targeted Activations: Input-Only Suppression of Evaluation-Awareness Latents in Large Language Models
- Toward Standardized Cross-Vendor Agent Tool Trust Management in Autonomous Networks
- From Role Prompt to Infinite Thinking: Exploiting Persona Conditioning for Inference Cost Attacks in LLMs
- Falling Behind Drives Unsafe Development in an Idealised AI Race Experiment
- "Uncensored" open LLMs are measurably more optimistic than their base models
- DONDO: Open w2v-BERT Speech-Recognition Base Models for African Languages
- Agentic Evaluation of Copyright Law Compliance
- Nanbeige4.2-3B: Unlocking Agentic Capabilities in a Compact Model
- TRACE-ROUTER: Task-Consistent and Adaptive Online Routing for Agentic AI
- Opaque Epistemic Mediation: How LLM Deployment Configurations Shape the Validation of Pseudo-Science
- Semalith v1.4: A Calibrated 184M Safety Classifier Achieving State-of-the-Art Prompt-Injection Detection at 44x Fewer Parameters than Llama-Guard-3-8B
- AIR-BENCH Live: An Evolving Safety Benchmark for Foundation Models
- Towards Nexus-Score: Metadata Gaps Limit Scholarly AI Attribution
- PANOPTICON: A PII-Based Assemblage of Naturalistic Output Tokens for Investigating Privacy Leakage Within LLM Context Window
- StepX-Edge: An On-Device UI Vision-Language Model via Architecture-Training-Deployment Co-Design
- Not All LLM Reasoning is Visible in the Chain-of-Thought
- Beyond Exact Match: How Evaluation Methodology Dominates Model Choice in LLM-Based Product Attribute Extraction
- Mask2Shield: Strengthening LLM Safety against Neuron-Pruning Attacks
- False Prophets: On the Security of World Models in Agentic Systems
- Hiding in Plain Sight: An Effective Physical Adversarial Patch Attack against Visual-Infrared Fused Face Detection
- UltraViT: Latency-Optimized On-device Vision Encoder for Large Vision-Language Models
- Confidently Wrong: Exception Chain Collapse in Frontier LLM Rule Evaluation
- Inference-Time Consensus for Mitigating Hidden Behaviors from LLM Fine-Tuning
- Separating Capability from Permission: A Governance Framework for Agentic AI Autonomy Levels
- Isolated but Exposed: Persistence-Based Memory Extraction Attack on LLM Agents
- Do LLMs Know Their Vulnerable Scenarios?
- The Illusion of Secure LLM Code: Closing the Security Gap via Iterative Reprompting
- Melo: A Production LLM-Powered Music Recommendation Agent
- AI Strategy: How to Choose What AI Product to Implement
- N₀-VTLA: Scaling Vision-Tactile-Language-Action Model with Latent Tactile Tokens
- N₀-TWAM: Scaling Tactile-Native World-Action Model for Contact-Rich Manipulation
- Memory Efficient Audio Synthesis with Decoupled Temporal Depth Diffusion Transformers
- SpecBox: Speculative Sandbox Scheduling for Efficient LLM Agent Serving
- A Cybersecurity MLPS Large Language Model with Multi-Path Retrieval Fusion
- Just Testing, Move Along: Evasion of LLM-based System Log Interpretation by Prompt Injection
- EXE-Bench: Ranking the Tradeoffs of AI-based Windows Malware Detectors for Real-World Usability
- Face Age Verification Vulnerabilities Under Simple Appearance Manipulations
- Gubernaut: A Deterministic Homeostatic Controller for Affect-Regulated LLM Agents, Validated Across Independent Model Families
- Beyond Aggregate Risk: Role-Stratified Conformal Risk Control for LLM Tool Calls
- When LLM Defenses Backfire: Characterizing Safety, Performance, and Cost Trade-offs
- The SpiNNaker2 chip: a many-core platform for flexible and scalable brain-inspired computing
- LLM-SoccerArena: Benchmarking LLMs on Real-World Predictions in Sports
- Agentic Permissions Policy Algebra for Taint Confinement in LLM Agents
- RecGPT-V3 Technical Report
- CARE: Pre-Execution Command Verification for Shell-Executing LLM Agents
- Pixels for Programs? A Cross-Provider Case Study of Input-Token Accounting for Source Code as Text and Images
- Oxygen-TryOn: Fashion-Native Foundation Model for Any-item Virtual Try-On
- Every Model Cheats: Prompt-Level Mitigation of Cheating on Offensive Cyber Tasks
- Protocol-Level Attacks on Agentic Commerce Platforms: A Cross-Platform Taxonomy, AIP-Bench, and Unified Defense
- ToolGuardian: Declarative Security for AI Agent-Tool Interactions
- Certified in Theory, Broken in Practice: Assumption Gaps in Cryptographic Model Certification
- ISPCloak: Weaponizing ISP for Optimization-Free Physical Camouflage against Deepfake Detectors
- RIS-Kernel: A Model-Agnostic Architecture for Long-Context LLM Inference via Sparse Attention
- SIREN (Luring LLMs onto the Rocks): PAIR-Driven Preference Manipulation in Web-RAG Recommenders
- Efficient Online LLM Watermark Detection via Rao-Blackwellized E-Processes
- Analysing Self-Harm Representations in Language Models: a Cross-Architecture Study
- Zero-Shot Mission-Level Evaluation for Aerial MLLM Agents
- MEUSLI: a Multilingual Projector for LLM-based ASR and Beyond
- Universal BCI Personalization: One API for Frozen EEG Trunks and Foundation Models
- Dynamic Capability Scoping for Enterprise AI Agents: A Synthetic Dataset and Three-Source Permission Architecture
- Where FactsGo Missing: A LayerwiseTaxonomy and Per-Layer Attribution of Information Omissionin Air-Gapped LLM Agent Pipelines
- Compile, Then Page: Executable SOP Programs and a Capability-Gated Runtime for Procedural LLM Agents
- Making Open-Source Text LLM Watermarks Durable Against Merging
- Routing Subspaces: Auditing Evaluation-to-Deployment Mismatch in Fine-Tuned Language Models
- TopoGuard: Graph Theory Based Defenses Against Split-Knowledge Attacks on RAG
- Answer-then-Edit: Reasoning Skeleton Editing for Anti-Distillation with Preserved Utility
- Confidently Deceptive: How Confidence Amplifies the Risk of LLM Deception
- Domyn-Small: A European 10B Reasoning Language Model
- Marking the Wrong Symptoms: Evaluating LLM Watermarks in Medical Texts
- InferenceBench: A Benchmark for Open-Ended LLM Inference Optimization by AI Agents
- Incomplete Prompt Jailbreaks in Large Language Models
- PhantomFill: When the Form Demands an Answer, Language Models Invent One
- CANN Bench: Benchmarking Agent Generated Kernels against Real NPU and Algorithmic Limits
- Autonomous disproofs of the sum-product conjecture over R with GPT-5.5 Pro
- Evaluating and Guarding Citation Faithfulness in Agentic Scientific Synthesis
- PromptPack: Scaling LLM Annotation Agents for Online Recommendation
- StrideDiffusion: Accelerating Diffusion Models for Time-series Generation
- Geometric Configurations of Perturbed Jailbreak Prompts
- Frontier Financial Judgement: Can agents tell what might move a stock?
- Security Vulnerability Patterns in AI-Generated Code: A Cross-Model Comparative Study
- Leaky Language Models: Stealing Architecture and Inference Optimizations via Per-Token Timing
- IssueTrojanBench: Benchmarking AI Coding Agents Against Malicious Issue Requests
- Robostral Navigate
- Refusal-Gated Decoding: Preserving Refusal Behavior Under High-Temperature Sampling
- Which Model Is Actually Serving You? IRIS: Budgeted Black-Box Auditing of Model Substitution and Routing Dilution in LLM Gateways
- GeoThreat: Transferable Targeted Adversarial Attacks on Large Vision-Language Models for Remote Sensing Image Interpretation
- QuantiBias: Benchmarking Quantization-Induced Bias in LLMs
- V-DEAL: Diagnosing Video Safety De-Calibration as an Understanding-Refusal Coupling Failure
- Detectors Learn the Wrong Thing: Shortcut-Resistant Adversarial Training Against Physically Realizable Attacks
- The Dark Room in the Reward Channel: Dense Prediction Rewards Collapse GRPO-Trained LLM Agents -- and What Actually Works
- Emergent Misalignment Recruits a Pre-existing Persona Subspace
- Agentic Context Management: Solving Agent Memory and Cost by Treating Them as Lifecycle and Architecture Problems
- Same Dangerous Objective, Opposite Advice: Direct Exposure versus Multi-Agent Mediation
- AXIS: A Growable Community-Driven Data Engine for Scalable Robot Manipulation
- Coercion and Deception in AI-to-AI Management: An Agentic Benchmark of Unprompted Escalation
- Benchmarking Confidential GPU Inference on NVIDIA H100 under Intel TDX
- NEXUS: Structured Runtime Safety for Tool-Using LLM Agents
- Economic Evaluations of Language Models
- Predictive single cell foundation model for gene regulation and aging with privacy-preserving tabular learning
- ChannelGuard: Safe Models Do Not Compose into Safe Multi-Agent Systems
- ChainWatch: A Kill Chain-Aligned Sequential Detection Framework for Multi-Step Attacks in MCP-Based AI Agent Systems
- The Chronos Vulnerability: A Taxonomy of Temporal Persistence and Memory-Based Deception in Agentic AI
- Guardrails as Scapegoats: Auditing Unfaithful Safety Refusals in Tool-Augmented LLM Agents
- When HTTP 402 Meets the Blockchain: Risks on Emerging x402 Payments
- Twin Agent: Context Residual Compression for Privilege Separated Agents
- SafeGen: Goal-Conditioned Video Diffusion of Safety-Critical Scenarios for VLM-Based Autonomous Driving
- OPIUM: Mitigating Steering Externalities and Over-Refusal via Dual Objective Latent Optimization
- Know Your Agent: Reconnaissance-Driven Pentesting of AI Agents
- Defense Against LLM Backdoors using Critical Neuron Isolation Pruning
- WearWow: Native 2K Multi-Garment Virtual Try-On via Adaptive Token Packing and Preference Alignment
- HijackKV: New Threat in Position-Independent KV Cache Reuse
- SLAI T-Rex: Full-Parameter Post-training of the DeepSeek-V4 Family on Ascend SuperPOD
- The Ethics of Autonomous AI Agents for Offensive Security
- Sound Probabilistic Safety Bounds for Large Language Models
- Don't Trust the Label: License Laundering in AI Supply Chains
- PyroDash: Cost-Efficient Token-Level Small-Large Language Model Collaborative Inference
- Generative AI floods and dilutes the market for books